search

kabachuha / sd-webui-text2video

Auto1111 extension implementing text2video diffusion models (like ModelScope or VideoCrafter) using only Auto1111 webui dependencies
Other
1.28k stars 108 forks source link

Improved handling of different models #193

Closed DrMWeigand closed 1 year ago

DrMWeigand commented 1 year ago

Selecting base model is automatically done by drop-down selection and different types of modelscope models can be selected by parsing the subfolders in ModelScope/t2v/*

image

kabachuha commented 1 year ago

Tbh, I'm doing away from models/ModelScope/t2v location in favor of models/text2video/*, and the thing is legacy which is to be removed soon. (Add a commandline notice of need to relocate the model?)

DrMWeigand commented 1 year ago

This solution could be easily adapted to a new location of the models and would provide much easier switching between different models than offered currently